Paul H. Patterson - US grants

A number of neural crest derivatives have been shown, by manipulation of developmental signals in cell culture, to display remarkable plasticity in the range of phenotypes they can assume. The interconversion of cell types observed in culture has been further shown to reflect differentiation choices made by certain derivatives during normal development in vivo. It is proposed to identify the precursors and intermediate cell types that give risk to the various members of the sympathoadrenal lineage, to characterize the cellular and hormonal signals that control these differentiation decisions, and to determine the role of these signals in vivo. I. A protein secreted by heart cells can convert postmitotic, noradrenergic neurons into cholinergic neurons. we have purified this molecule to homogeneity and sequenced its 11 N-terminal amino acids. Antisera against a synthetic peptide of this sequence precipitate the cholinergic factor and directly block its activity. The antibodies and the peptide will be used to perturb the development of cholinergic neurons ia vivo, as well as to devise a RIA for localization of the factor in vivo. Oligonucleotide probes corresponding to this sequence are being used to screen heart cell cDNA libraries in an effort to clone the gene for this protein. Successful cloning would enable us (i) to use an expression system to produce large amounts of the protein for in vivo injections and chemical characterization, and (ii) to examine the possibility that there could be a gene family of such differentiation factors. II. Our previous work on the interconversion of adrenal chromaffin cells, small intensely fluorescent (SIF) cells and sympathetic neurons led too the hypothesis that SIF cells might be the central intermediate cell type of the sympathoadrenal lineage. Our new findings suggest that the putative precursor cell is not a mature SIF cell, but a cell that shares certain markers with both SIF and chromaffin cells. It is proposed to isolate these putative precursors in culture in order to determine what derivatives they can give rise to and what the nature of the signals are that influence these phenotypic choices. The relationship of these precursors of neural crest cells will be probed through the production of new lineage markers using our immunosuppression technique.

A variety of cellular interactions in the developing nervous system are thought to be mediated by macromolecules which neurons have on their surfaces or which they secrete into the extracellular space. Recent work with cultures of dissociated sympathetic neurons has described a number of candidates for such intercellular signals. These are surface-bound and secreted molecules, identified by isotopic labeling or with monoclonal, antibodies, which are enriched in either adrenergic or cholinergic sympathetic neurons. Such molecules are thus potentially involved in functions or interactions specific to one of these two phenotypes. Since certain sensory neurons are known to exert trophic influences on their targets it is proposed to examine neuronal cultures from various sensory ganglia with the same methods to see if different classes of sensory neurons also secrete characteristic families of glycoproteins. The functional roles of the glycoproteins and glycolipids of interest are being investigated by producing antibodies against them. The antibodies are then added to cultured sympathetic neurons, either chronically or in acute incubations, and the effect on neuronal survival, growth and differentiation determined. In such experiments, certain of the monoclonal antibodies already produced against sympathetic neuron surface membranes of have been found to selectively decrease or enhance the development of specific neuronal properties such as transmitter synthesis or release. The new monoclonals to be generated agaist phenotype-specific molecules as well as the rest of the present monoclonal library (greater than 100 clones) will be similarly examined for effects on neuronal development. The antibodies will also be tested for their ability to disrupt previously studied cellular interactions such as neuron-induced receptor clustering on myotubes, and Schwann cell mitosis induced by contact with axons. Another method being employed for assessing antigen function is to use the antibodies to generate mutant PC12 cell lines which are then tested for deficiencies in the various assays. Selected antigens will be localized in vivo by immunocytochemical methods and, ultimately, antigen function in normal development will be assessed by antibody injections in developing rats.

Recent work from the laboratory of Dr. Patterson as well as from other research groups has identified several secreted and cell surface molecules which may be involved in the intercellular interactions mediating axon growth and synapse formation. The goal of this project is to determine the functional role in neuronal development and regeneration of one of these sets of molecules . proteolytic enzymes or proteases. Dr. Patterson and his colleagues have biochemically characterized several proteases, including plasminogen activator, which are spontaneously released by the distal tips of growing neurites in culture. In addition, these researchers have partially characterized two irreversible plasminogen activator inhibitors which are released by muscle and peripheral glial cultures. Recent evidence from Dr. Patterson's lab has shown that inhibition of plasminogen activator can increase the rate of neurite outgrowth in culture. The proposed work will extend these studies to an in viv o (rat and chick) preparation. The experimental strategy is to use antibodies to perturb the activity of this protease and observe the effect on regenerating peripheral and central nerves in the intact animal. The establishment of complex synaptic arrays during vertebrate neural development is the result of an equally complex series of cellular interactions between neurons and their targets. These interactions occur throughout the various stages of cell migration, axonal outgrowth, recognition of target cells, differentiation of synaptic elements, and reorganization and maintenance of final connections. This research will contribute to our understanding of the role of proteases in axonal outgrowth during development and regeneration in the nervous system.

The monoclonal antibody technique enables the production of highly specific reagents using small amounts of heterogeneous material. The reagents can be used to purify molecules or cells of interest, localize antigens in the animal, and probe their functions. The Caltech Monoclonal Antibody Facility provides staff, equipment, and expertise to produce hybridomas for investigators in the Biology Division. The facility has been successful in producing monoclonal antibodies, and new discoveries and fundamental advances are likely to be forthcoming. This award will provide support to continue operation of the facility and to increase its productivity.

A key, but insufficiently understood area of neural development is the molecular basis of neurospecificity. At the physiological level, one of the best studied examples of synaptic specificity in vertebrates is the rostrocaudal, positional bias displayed by spinal cord axons as they innervate sympathetic ganglia or transplanted intercostal skeletal muscles. In an effort to examine the molecular basis of this positional specificity, we used the cyclophosphamide immunosuppression method to produce of monoclonal antibody (mAb), ROCA1, which binds preferentially to rostral ganglia and rostral intercostal nerves. ROCA1 finds specifically to a 6OkD and a 26kD protein. We have identified the former antigen as the neuron-specific, intermediate filament, peripherin, and immunoblots demonstrate that peripherin is expressed in a graded manner in intercostal nerves, declining in caudal segments. The 26kD antigen is a surface membrane protein expressed by neurons and glia. ROCA1 binding to the 26kD protein in sections of nerves and ganglia declines dramatically in caudal segments, apparently due to a masking of this epitope. The proposed research will extend these findings in four directions. (i) The 26kD protein has been purified for amino acid sequencing. This data will be used to isolate the corresponding cDNA clones. Cloning will allow comparison with sequences of known proteins and functional motifs, (ii) Results with likely anti-idiotype Abs suggest that the 26 and 6OkD proteins may interact with each other. This will be further explored using the anti-idiotype Abs and direct binding studies. (iii) The cell- and position-specific expression of both peripherin and the 26kD protein is being explored in rat embryos with several Abs. (iv) Expression of peripherin and the 26kD protein is being examined in cell culture. Questions under study include: is peripherin preferentially expressed by rostral neurons in culture and, if so, does that difference require association with glia; does the apparent masking of the ROCA1 epitope on the 26kD protein in caudal glia depend on association with caudal neurons.

During the past grant period we used amino acid sequences from the cholinergic differentiation factor (CDF) to clone the rat cDNA for the cytokine also known as leukemia inhibitor factor (LIF). Using CDF/LIF and other recombinant proteins, and a novel cell culture assay, our work contributed to the recognition of the neuropoietic cytokine family. We continue to look for novel members of this family and have begun to explore the in vivo role of LIF in development, in the response to neural injury, and in the synaptic plasticity involved in LTP. LIF: (1) Our study of LIF-mutant mice reveals atrophic changes in the hippocampus and the visual cortex. These observations will be extended to the examination of brains from developing animals and the use of many other phenotypic markers. We have also obtained embryonic brains of LIFR-mice, and brains of rats that were injected with LIF-blocking antibodies at birth; the rats display behavioral abnormalities. In complementary experiments, we will inject LIF-secreting fibroblasts into the brains of normal and LIF-mice of various ages. (2) To determine if LIF acts directly on CNS neurons, it will be tested on cultured hippocampal and cortical neurons for effects on survival and gene expression. (3) We find that LIF expression is strongly induced in glial cells upon nerve injury, and that the neuropeptide induction that normally follows nerve section is much reduced in LIF-mice. We will now examine whether neuronal survival and the rate of nerve regeneration are altered following nerve section in LIF-mice. We will also investigate whether IL-1 mediates the LIF induction following injury in vivo. (4) We have begun to explore the role of LIF in neurogenic inflammation. We find that LIF is induced in the skin by intradermal CFA, a paradigm that leads to neuropeptide induction in sensory neurons. We will test whether this neuropeptide induction also occurs in LIF-mice subjected to CFA. (5) Since LIF and its receptor mRNAs are found in the hippocampus, and LIF mRNA is elevated by high levels of activity, we will ask if this protein has a role in LTP. The effects of LIF and its blocking antibody on synaptic physiology in hippocampal slices will be monitored, and LIF mRNA will be assayed following LTP. Preliminary results indicate that LIF suppresses the induction of LTP in slices. New Cytokines: (1) We are searching for homologues of GPA, CNTF and LIF. Bands of the predicted sizes have been obtained using the PCR for all three cytokines in chick, mammal and fish, and these are being characterized. Northern analysis, cloning full cDNAs and protein expression are planned. (2) Since another neuropoietic cytokine, OSM, is only available from human, very little is known about its potential role in the nervous system. We are attempting to clone the mouse OSM gene using nested PCR primers, and would like to examine OSM expression in the normal and injured PNS and CNS.

We have produced monoclonal antibodies (mAbs) that recognize candidate downstream targets of homeobox transcription factors. The antigen for one of the mAbs, FORSE-1, is the focus of this project. FORSE-1 binds the surfaces of proliferative neural precursor cells in the embryonic forebrain. It's pattern of labeling delineates a novel set of boundaries in the rodent telencephalon and diencephalon, displaying a partial overlap with the expression of the transcription factor BF- 1. We have identified the FORSE-1 epitope as the oligosaccharide Lewis x (Le x), and the antigens that carry this epitope in the brain as neutral glycolipids and the chondroitin sulfate proteoglycan, phosphacan. We propose to investigate the function of Le x in vivo and in vitro. (1) Migration in vivo We have initially focused on the perturbation of neuronal migration in vivo by injection of the FORSE-1 mAb, control mAbs and synthetic Le x oligosaccharide antagonists. Injection of the synthetic antagonists or FORSE-1-secreting hybridoma cells into the ventricle of E15.5 rat embryos severely disrupts the migration of BrdU-labeled neurons to cortical layers 5 and 6. Injection of hybridoma cells secreting the HNK-1 mAb, in contrast, has no effect on migration of these cells. Additional positive and negative control mAbs and oligosaccharides will be tested in this in vivo system. (2) Migration in vitro In order to elucidate further the mechanism by which migration is inhibited, both the oligosaccharide perturbants and the mAbs will be used in a slice culture system in which the migration of BrdU- or DiI-labeled neurons are followed (using time-lapse imaging in latter case). (3) Axon guidance The in vivo perturbation procedure will also be used to test a role for LeX in axon guidance. Cortical and retinal axon tracts in mAb- or oligosaccharide-injected embyros will be labeled with TAG-1 and NF mAbs, and with DiI. Size of tracts, trajectory of growth cones and degree of fasciculation will be measured. (4) Fuc-T knockouts We have obtained mice in which two of the five known fucosyltransferases (Fuc-Ts) that synthesize Le x have been disrupted by homologous recombination. The brains of these knockout embryos will be analyzed for FORSE-1 labeling and, if negative, for defects related to neuronal migration and axon outgrowth in vivo and in vitro. If the available mice are not deficient in brain Lex , we will collaborate on cloning and disrupting a brain-enriched form of Fuc-T.

DESCRIPTION (provided by applicant): Huntington's disease (HD) is a fatal neurodegenerative disorder caused by the expansion of CAG repeats in the huntingtin (Htt) gene. Mutant Htt forms intracellular aggregates and is cytotoxic to specific neurons in the striatum and cortex. Causes of cell death in HD include the activation of cell death genes and/or alteration of normal transcription. The objective of this proposal is to investigate how mutant Htt interacts with NF-kappaB signaling pathway. This pathway regulates expression of pro-survival genes, is central to the function of neurotrophic factors and cytokines, and is activated during excitotoxic cell death mediated by NMDA receptors. We are investigating the interaction of mutant Htt with the IkappaB kinase (IKK) complex that regulates NF-kappaB, and how this affects neuronal responses to extracellular signals. We will focus on the mechanism of how the direct interaction between mutant Htt and IKKgamma, a regulatory component of IKK, influences neurodegeneration in HD. Molecular, cell culture, brain slice and animal model approaches will help to explore the nature and consequence of this interaction. We have preliminary evidence that IKKgamma influences NF- kappaB-regulated neuronal gene expression, as well as mutant Htt-induced cell death and aggregation. (i) Using in vitro and cellular assays, we are mapping the domains in mutant Htt and IKKgamma that mediate their binding. (ii) The functional consequences of Htt-IKKgamma interaction are assessed by isolating IKK complex from mutant and normal Htt-expressing cultured cells and HD transgenic mouse brain. Its kinase activity and ability to regulated NF- kappaB mediated gene expression is then tested using in vitro kinase and gene reporter assays. (iii) Downstream genes influenced by the Htt-IKKgamma interaction will be identified by cDNA mini-arrays. mRNA will be isolated from several brain regions from HD and normal mice. (iv) The influence of IKKgamma on mutant Htt-induced cell death and aggregation is being examined in HD transgenic mice and in PC12 cells that inducibly express Htt and/or IKKgamma. We have constructed lentiviral vectors expressing a dominant negative form of IKKgamma as well as cell permeable peptides that block IKKgamma activity in cells. These reagents will be injected into the brains of HD mice in tests of their efficacy in inhibiting mutant Htt toxicity and aggregation. They will also be tested on brain slices from HD mice, in the presence and absence of NMDA.

DESCRIPTION (provided by applicant): Maternal viral infection can increase the incidence of mental illness and cerebral palsy in the offspring. Such infections can also lead to miscarriage, premature and stillbirth and early neonatal mortality. We are utilizing a mouse model to examine the effects of maternal respiratory infection with influenza virus on fetal brain development. We find that the adult offspring of these mothers exhibit interesting behavioral abnormalities and striking responses to psychoactive drugs, as well as a neuropathological finding that is remarkably similar to that found in autism. (i) We have preliminary evidence that adult mice born to infected mothers exhibit a highly localized loss of Purkinje cells, similar to that observed in autistic brain. We will quantitate Purkinje cell density, as well as the width of the molecular layer in this and other parts of the cerebellum, in both adults and during development. (ii) It is possible that the infected mothers have altered maternal behavior, which could have lasting effects on the offspring. To investigate this, we will carry out cross-fostering experiments, in which the offspring of infected mothers are raised from birth by control mothers. These offspring will be tested behaviorally with the assays we have employed previously, and their brains will be examined by histology. (iii) As we have not detected viral infection of the fetus, it is possible that these morphological and behavioral abnormalities found in the offspring are due to the anti-viral inflammatory response mounted by the mother. Supporting this hypothesis are results of injecting pregnant mice with poly I:C, a double stranded RNA that causes an inflammatory response resembling that induced by viral infection. At least one of the behavioral abnormalities in the offspring can be elicited by this procedure, so further analysis of such offspring is merited. We also plan to test the effects of viral vaccination of pregnant mice on the behavior and neuropathology of the offspring. (iv) In parallel studies using cytokine knockout mice and cytokine adenoviral vectors, we are able to manipulate the inflammatory response in the brain and periphery. These and other strategies will be used to test whether manipulation of the mother's response to viral infection can prevent the development of brain abnormalities. Importantly, anti-cytokine strategies in influenza-infected mice have already proven effective in treating fever and lung inflammation. We will also use cytokine injections of non-infected mothers to help identify cytokines that may mediate the effects of maternal viral infection.

The cytokine leukemia inhibitor factor (LIF) can regulate neuronal birth, survival and phenotype in cell culture. Moreover, we have found that endogenous and exogenous LIF control neurogenesis in the adult olfactory epithelium following injury in vivo. We are now using LIF knockout (KO) mice and LIF delivery by viral vectors to evaluate the hypothesis that LIF can control neurogenesis in the adult CNS in vivo. We find that viral delivery of LIF, but not LacZ, into the lateral ventricle stimulates BrdU labeling in the subventricular zone (SVZ), in the nearby hippocampus, cortex and striatum, as well as in the olfactory bulb (OB). Sacrificing animals at various times after BrdU injection shows that the labeled cells in the OB come from the SVZ, suggesting that the exogenous LIF is stimulating the production of OB granule neurons. Conversely, we find that eliciting seizure stimulates BrdU labeling in the hippocampus in wild type (WT) but not LIF knockout (KO) mice. Both of these observations will be examined further. LIF regulation of neurogenesis in the OB: (i) We will follow the fates of the BrdU + cells that migrate into the OB following injection of the LIF adenovirus into the lateral ventricle using histochemical markers. (ii) We will ask whether endogenous LIF plays a developmental role in the OB using LIF KO and WT mice and histochemical markers. (iii) To clarify interpretation of our results with LIF delivery by the adenoviral vector, we are making lentiviral vectors for comparison. LIF regulation of neurogenesis in the hippocampus: (i) We will ask if endogenous LIF controls seizure-induced neurogenesis in the hippocampus using BrdU and neuronal markers. Early results indicate many fewer BrdU+ cells in the LIF KO hippocampus than in the WT following seizure. (ii) Our experiments with lateral ventricle injection of adenoviral vectors show a remarkable stimulation of BrdU labeling in the hippocampus induced by LIF, but not LacZ. Cell-specific markers will be used to identify these proliferating cells. Analysis will be done at several stages following viral injection so as to be able to follow the fates of the cells over time.

DESCRIPTION (provided by applicant): Huntington's disease (HD) is a fatal neurodegenerative disorder caused by expansion of the polyQ domain in the huntingtin (Htt) protein. Mutant Htt protein forms intracellular aggregates and is cytotoxic to specific neurons in the striatum and cortex. We generated monoclonal antibodies (mAbs MW1-8) that bind the epitopes polyQ, polyP or the C-terminus of exon 1 in Htt. To use these mAbs as intracellular perturbation agents, we cloned their antigen-binding domains as recombinant, single chain Abs (intrabodies). In 293 cells co-transfected with expanded polyQ Htt, MW7 intrabody, recognizing the polyP domains of Htt, significantly inhibits aggregation as well as the cell death induced by mutant Htt. In contrast, MW1 and 2 intrabodies, recognizing the polyQ domain, stimulate Htt cell death. These results suggest MW7 as a candidate for gene therapy in HD. We propose to explore the therapeutic potential of MW7 and several new anti-Htt intrabodies in several models, and to produce new, more efficacious intrabodies. (i) Preliminary results indicate that MW7 is effective in an in vivo-like system: co-transfection of MW7 with mutant Htt in acute striatal slices inhibits Htt neurotoxicity. Further experiments are required to support this result, (ii) To engineer more potent intrabodies, we screened a naive human scFv phage library for binding to a proline-enriched peptide from Htt, followed by a screen for preferential binding to mutant Htt. Six new intrabodies were cloned and two of these appear very promising in blocking mutant Htt aggregation and toxicity in cell culture, (iii) Three animal models are being investigated in therapeutic experiments. First, a lentiviral vector encoding mutant Htt is injected in the adult striatum, causing neuropathology. Co-injection of vectors encoding the intrabodies will be tested for blocking this pathology. Second, AAV vectors encoding the best candidate intrabodies are injected in neonatal the R6/2 HD mouse striatum. These mice will be examined as adults for alterations in Htt aggregation, neuropathology and HD-like motor symptoms. Third, the AAV-intrabodies are injected into the adult striatum of pre- and post-symptomatic R6/2 and N171 HD model mice. These mice will be examined as above, (iv) To identify novel Abs with therapeutic potential we are developing a new screen for intrabodies that enhance the survival of PC12 cells induced to express mutant Htt. This non-biased approach could uncover new pathways involved in HD pathogenesis.

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. Primary support for the subproject and the subproject's principal investigator may have been provided by other sources, including other NIH sources. The Total Cost listed for the subproject likely represents the estimated amount of Center infrastructure utilized by the subproject, not direct funding provided by the NCRR grant to the subproject or subproject staff. Based on epidemiological evidence that viral infection of pregnant women can strongly increase the risk for autism in their offspring, a mouse model of maternal immune activation has been established. Pregnant mice are given a respiratory infection at midgestation, and the offspring display a series of behavioral abnormalities consistent with those seen in autism, including deficits in social interaction and investigation of novel objects, as well as increased anxiety under mildly stressful conditions. Moreover, these offspring exhibit a specific type of neuropathology that is often seen in autism: spatially localized deficit of Purkinje cells in the cerebellum. Now that the mouse model has shown behavioral and histological abnormalities consistent with those found in autism, it is important to extend this work to primates. These animals possess a behavioral repertoire, such as the expression and interpretation of facial expressions, which is very homologous with human social behavior. Non-human primates therefore provide a sensitive indicator of impairments in social behavior and communication. A valid primate model of autism would also provide a unique resource for evaluating therapeutic interventions.

Mobilizing endogenous progenitor cells in the adult brain for treatment of neurodegenerative and demyelinating disease offers several advantages over transplantation of stem cells (SCs). The present project is investigating the ability of the cytokine leukemia inhibitory factor (LIF) to stimulate SC and progenitor cell production, and to direct their differentiated fates in a demyelination model. Since LIF is up- regulated in response to a variety of human brain injuries and diseases, experimentally increasing the level of LIF in the adult brain serves to enhance the normal response mechanism. Elevating LIF in the adult mouse brain promotes the self-renewal of NSCs, thereby increasing the number of cells that may be available for repair. In addition, LIF stimulates oligodendrocyte progenitor cell (OPC) proliferation and promotes oligodendrocyte survival. Thus, LIF can enlarge the NSC pool as well as direct NSCs towards fates that are of potential clinical importance. (1) It is proposed to examine the efficacy of this cytokine in the cuprizone model of demyelination, in which the onset of de- and remyelination can be experimentally controlled. Preliminary results indicate that exogenous LIF can significantly increase the number of OPCs and oligodendrocytes following demyelination. (2) Preliminary results also indicate that remyelination is strongly improved by LIF treatment. To explore the possibility of LIF-driven functional recovery, a collaboration has been established to study axonal conduction. (3) How does LIF increase OPCs, oligodendrocytes and myelin? Does it act directly on each of these cell types to direct their proliferation and differentiation? Preliminary results suggest that LIF acts directly on both newly generated OPCs and oligodendrocytes. Since LIF also activates glial cells, which in turn may influence OPCs, we will ask whether LIF stimulates OPC proliferation by direct signaling through its receptor in OPCs by conditionally deleting the gp130 component of the LIF receptor in OPCs. Revised Specific Aims (modified for 2-year funding) Mobilizing endogenous progenitor cells in the adult brain for treatment of neurodegenerative and demyelinating disease offers several advantages over transplantation of embryonic or neural stem cells (NSCs). We are investigating the ability of the cytokine leukemia inhibitory factor (LIF) to stimulate stem and progenitor cell production, and to help direct their differentiated fates following demyelination. Since LIF is up-regulated in response to a variety of human brain injuries and diseases, increasing the level of LIF in the adult brain by injecting the recombinant protein, or viral vectors encoding it, serves to enhance the normal response mechanism. During the prior granting period, we found that elevating LIF promotes the self-renewal of NSCs in the adult brain subventricular zone (SVZ), thereby increasing the number of cells that may be available for repair. In addition, we showed that LIF stimulates oligodendrocyte progenitor cell (OPC) proliferation in the brain, and promotes oligodendrocyte survival following spinal cord injury. Finally, we found that LIF enhances the replacement of injured adult olfactory sensory neurons. Thus, LIF has the ability to not only enlarge the NSC pool, but also to direct neural stem and progenitor cells towards fates that are of potential clinical importance. We propose to explore these properties of LIF a mouse demyelination model. A1. Test the ability of LIF to increase OPCs and oligodendrocytes following demyelination. The congenital and acquired CNS demyelinating diseases of children and adults are attractive targets for cell replacement therapy because the primary cell loss is of a single cell type, the oligodendrocyte. We found that exogenous LIF increases PHS 398/2590 (Rev. 11/07) Page 7 Continuation Format Page Patterson, Paul H the number of OPCs in the normal mouse brain as well as mature oligodendrocytes in injured spinal cord. Therefore, we are examining the efficacy of this cytokine in the cuprizone model of demyelination, in which the onset of de- and remyelination can be experimentally controlled to determine whether LIF can promote oligodendrocyte generation in the context of demyelination. We will deliver LIF protein or a viral vector encoding LIF and examine OPC proliferation, as well as the number of newly generated, mature oligodendrocytes using cell stage-specific markers coupled with BrdU administration. Preliminary results indicate that exogenous LIF significantly increases the number of OPCs and oligodendrocytes following cuprizone-induced demyelination. Hypothesis: Delivery of LIF can stimulate OPCs and restore oligodendrocyte numbers to normal levels following demyelination. A2. Test the efficacy of LIF to support remyelination and restore function. In light of these encouraging preliminary results, we have begun to examine remyelination in the cuprizone model. First, we are quantifying the extent of myelination using immunohistochemistry for myelin proteins. Second, we are assessing recovery of the characteristic features of nodes of Ranvier by assaying for markers of nodal, paranodal, and juxtaparanodal structures. Third, we are using electron microscopy (EM) to examine the ultrastructural features of the newly formed myelin and to quantify myelin thickness and the percent myelinated fibers in the medial corpus callosum (CC). To explore the possibility of LIF-driven functional recovery, we have established a collaboration to study axonal conduction using electrophysiology. Hypothesis: LIF can stimulate remyelination and aid in functional recovery in the cuprizone demyelination model. A3. Determine the targets of exogenous LIF action and the range of its effects on those targets. How does LIF stimulate OPC proliferation, oligodendrocyte production and myelin formation? Does it act directly on OPCs to direct their proliferation and their differentiation? Or does LIF set in motion a train of events that is subsequently controlled by other factors? In the in vivo setting, we are first characterizing the progenitor cells that proliferate in response to LIF by combining BrdU and cell- and stage-specific markers with staining for phosphorylated STAT3 (pSTAT3), a reliable marker for detecting LIF activation of the signal cascade downstream of its receptor. Preliminary results suggest that LIF acts directly on both newly generated OPCs and oligodendrocytes. Second, since LIF also activates microglial cells and astrocytes, which in turn may influence OPCs, we will ask whether LIF stimulates OPC proliferation by direct signaling through its receptor in OPCs by conditionally deleting the gp130 component of the LIF receptor in OPCs. Hypothesis: LIF stimulates oligodendrocyte generation directly by activating gp130 signaling in OPCs. To facilitate completion of our aims in a two-year funding period, we have eliminated several non-essential experiments from Aim 1 and Aim 3. These changes are outlined below. Aim 1 modifications The majority of the experiments in this proposal take advantage of the cuprizone model, which is the best model for our proof-of-concept experiments. In our most recent proposal, we included an additional experiment using a chronic EAE demyelination model (D1, Expt. 3), the aim of which was to test the efficacy of LIF in stimulating OPC proliferation and oligodendrocyte generation in a model that more closely mimics the pathology observed in MS. While understanding whether the effects of LIF that we observe in the cuprizone model extend to an inflammatory demyelination model is important for an evaluation of the therapeutic potential of LIF for human demyelinating PHS 398/2590 (Rev. 11/07) Page 8 Continuation Format Page Patterson, Paul H disease, it is not necessary for our initial proof-of-concept experiment, and thus this experiment can be delayed until further funding is obtained. Aim 2 modifications We have made no modifications to this aim as we think these experiments are critical for our proof-of-concept experiments. Furthermore, our encouraging preliminary results, including both those reported in the progress report and those accumulated since submission, make us confident that we can complete all components of this aim within two years. Aim 3 modifications Two of the reviewers stated that our third aim was our most interesting aim, and we agree, but nevertheless several of the experiments in this aim are not essential for our initial proof-of-concept study since they shift the focus from the initial characterization of the potential therapeutic value of LIF delivery following demyelination to an exploration of the mechanism underlying LIF actions. First, we have eliminated the experiment in which we conditionally delete gp130 in PDGFRa+ OPCs following cuprizone treatment (D3, Expt. 2b). We still intend to conditionally delete gp130 in OPCs just prior to LIF treatment, which will be straightforward and will address our primary question: does LIF stimulate OPC proliferation directly through the activation of gp130 signaling in a cell-autonomous manner? Second, we have eliminated our experiment aimed at understanding whether the continued presence of LIF has additional effects on differentiating oligodendrocytes beyond the stimulation of OPC proliferation (D3, Expt. 3). We initially suggested that this experiment would be completed in year 4, and it is not currently high on our priority list. Eventually, it will be necessary know whether transient LIF treatment is sufficient for its therapeutic effects or whether long-term LIF delivery is required, but these experiments can be delayed until further funding is obtained. Finally, we have eliminated the analysis of demyelination and remyelination in LIF KO mice. While our preliminary findings of delayed oligodendrocyte generation are intriguing, demonstrating whether the cause of this delay stems from a reduced progenitor response, a decrease in survival of the newly generated oligodendrocytes, or any one of several other potential causes, which we outlined in our proposal (D3, Expt. 4), will involve the assessment of LIF KO and WT mice at multiple additional time points during cuprizone treatment. This would require a significant number of experimental animals and extensive tissue processing, immunostaining, microscopy and analysis. Therefore, without further assistance, completion of this experiment would be difficult. Removal of this aim does not undermine our overall objectives.

DESCRIPTION (provided by applicant): Despite the impact on society, and the demonstrated utility of animal models for understanding disease, relatively few researchers work on animal models of mental illness such as schizophrenia. A major stumbling block is, how does one know if a mouse is schizophrenic? Although it is possible to measure behaviors related to the negative symptoms of schizophrenia such as deficits in social interaction, cognitive flexibility and sensory-motor gating, it is said to be impossible to evaluate the positive symptoms of schizophrenia, hallucinations and delusions. Thus, this disorder is often considered to be "uniquely human". We aim to test this assertion by developing an assay for a positive symptom of schizophrenia. We have developed an animal model with relevance for schizophrenia that is based on epidemiological findings that the risk for this disorder is strongly increased by maternal infection. The offspring of mice given a respiratory infection at mid-pregnancy display deficits in a series of behaviors relevant to schizophrenia. These include social interaction, prepulse inhibition, latent inhibition, interaction with a novel object, as well as increased anxiety under mildly stressful conditions. Moreover, these offspring display neuropathology that is characteristic of schizophrenia. We propose a novel approach towards animal studies of a positive symptom of schizophrenia. This will require a major advance with broad implications: establishing functional magnetic resonance imaging (fMRI) in awake mice. Using surrogate histological measures of neuronal activity, it appears that the mouse brain responds to hallucinogens in a manner similar to the human. Moreover, preliminary experiments indicate that at least one strain of mice can be adapted to the restraining device, allowing fMRI scanning. If this procedure can be established reliably, we will test the offspring of infected mothers for sensitivity to subthreshold doses of hallucinogens, as well as for spontaneous hallucinations, as defined by activity in the visual or somatosensory cortices in the absence of sensory input. Even partial success in this project could open up several new fields of investigation, attracting investigators into the important area of mouse models of mental illness. Auditory hallucinations are reported by 50-70% of schizophrenia patients, and they are often associated with acts of violence and suicide. Since the cause is unknown, an appropriate animal model of hallucination-like brain activity would be extremely useful in exploring novel therapeutic avenues.

DESCRIPTION (provided by applicant): Huntington's disease (HD) is an autosomally inherited neurodegenerative disorder that is caused by expansion of CAG repeats in exon-1 of huntingtin (Htt). Enzymatic cleavage of mutant Htt protein leads to the generation of neurotoxic, amyloidogenic fragments and is one of the earliest events in HD pathogenesis. The environmental cues and the signaling pathways that regulate Htt cleavage are poorly understood, however. HD is an age-dependent disorder and age-related factors such as the accumulation of DNA damage and neuroinflammation are likely to influence its progression. The IkB kinase (IKK)/NF-kB signaling pathway is activated by these environmental factors and is implicated in the pathogenesis of HD. IKK? is responsive to DNA damage and mediates inflammatory responses. Our results demonstrate that pre- symptomatic HD mice have elevated IKK? activity localized to the striatum, while symptomatic HD mice also exhibit elevated IKK? activity in other brain regions including cortex. Inhibition of IKK? reduces the neurotoxicity of amyloidogenic fragments of Htt in a brain slice culture model of HD. We also find that IKK? activation induced by DNA damage promotes caspase-dependent WT and mutant Htt cleavage in post-mitotic human neurons. Genetic and chemical inhibition of IKK? prevents caspase activation and Htt cleavage, while increasing neuronal resiliency to DNA damage. Recent studies implicate IKK? in the phosphorylation of N-terminus (Ser13 and Ser16) of Htt. Phospho- mimetic modification of these residues ameliorates disease in HD mice. Preliminary studies from our laboratory and other investigators indicate that IKK? inhibition promotes the phosphorylation of Htt in neuronal models. Thus, IKK? is a key regulator of HD pathogenesis and its inhibition may be protective in HD. Moreover, deregulated IKK? is likely responsible for neuroinflammation, which occurs years before the onset of motor symptoms in patients. We propose to investigate the role of IKK? in HD pathogenesis in several animal models. We will delete IKK? in the brains of HD mice and examine whether the lack of IKK? prevents Htt cleavage, neuroinflammation and thereby ameliorates pathology. Furthermore, we plan to test the efficacy of small molecule inhibitors of IKK? as potentially novel therapeutics for HD. PUBLIC HEALTH RELEVANCE: Prior results with cultured cells described the role of the key kinase, IKKb, in the production of amyloidogenic, N-terminal mutant huntingtin, as well as in the induction of pro-inflammatory cytokines. These same inflammatory markers accumulate years before the onset of motor symptoms in Huntington's disease (HD) patients. The proposed specific aims will now evaluate the role of IKKb in mutant Htt neurotoxicity and neuroinflammation in mouse models of HD, and test the therapeutic potentials of small molecule inhibitors of IKKb.

DESCRIPTION (provided by applicant): Autism spectrum disorder (ASD) comprises a set of complex neurodevelopmental disabilities characterized by repetitive/stereotypic behaviors and deficits in communication and social interaction. Recent studies highlight striking neural and peripheral immune dysregulation in ASD. Moreover, a significant subset of ASD children exhibit gastrointestinal (GI) complications, including increased intestinal permeability and altered composition of intestinal microbiota. The potential connections between GI abnormalities, intestinal bacteria, and behavioral deficits have not yet been convincingly investigated. To examine the hypothesis that GI pathology is associated with, and contributes to behavioral symptoms, we employ a mouse model of an ASD risk factor, maternal immune activation (MIA). Our results show that these mice, which display cardinal ASD-like behaviors and neuropathology, also exhibit GI pathology. This includes changes in expression of tight junction components in the intestinal epithelium and a 'leaky gut', or diminished epithelial barrier function, which is reported in a significant subset of ASD children. Remarkably, this leaky gut is associated with an altered metabolite profile in the serum of the MIA mice, suggesting that GI permeability results in translocation of bacterial products into the circulation. Furthermore, we show that administration of a probiotic bacterium, Bacteroides fragilis, to these mice cures several behavioral abnormalities while restoring GI barrier function. Our central hypothesis is that correcting GI abnormalities with probiotic bacteria may be a safe and effective treatment for some of the abnormal behaviors in ASD. The specific aims that will test this hypothesis are: 1) in mechanistic experiments, determine if a cytokine relevant to MIA induces leaky gut; 2) determine whether putative metabolites that leak from the gut contribute to or modify behavioral abnormalities; 3) extending our results from the mouse model of an environmental ASD risk factor, test if this probiotic treatment corrects GI and behavioral abnormalities in genetic models of ASD; 4) determine if human ASD subjects with GI complications display altered serum metabolites. Based on compelling preliminary evidence, this project aims to explore the potential connection between GI barrier defects and altered behavior in preclinical models of autism, and extend these findings to humans. Our long-term goal is to explore possible serum biomarkers for ASD diagnosis, and potentially develop a novel probiotic therapy for at least a subset of children with ASD with GI issues.